Networks of computers are commonly used to implement distributed data processing systems. In particular, the INTERNET has attained a widespread diffusion in the last years. The INTERNET is a network of networks connecting millions of computers with a decentralized design. The INTERNET has a client/server architecture, wherein server computers of the network support shared resources; client computers access the shared resources through the network.
The INTERNET also allows each client computer to interact dynamically with a server computer, in order to download and display information whose content changes each time it is provided by the server computer. This function is typically used for handling generation and distribution of reports resulting from queries on a database of the server computer.
A solution known in the art for delivering dynamic information in the INTERNET consists in defining the reports by means of corresponding programs, for example conforming to the ActiveX scripting (ASP) specification. Whenever a request for dynamic information is received from a client computer of the network, the server computer invokes execution of the corresponding ASP program. The ASP program generates the report by running dynamic queries on the database of the server computer.
A drawback of the solution described above is that the programs generating the reports are quite complex; moreover, the dynamic queries run by these programs are relatively slow. As a consequence, the performance of the server computer is severely affected.
The above-mentioned drawback is particularly acute in a high-availability server computer, which is required to distribute complex reports at very high rate. In this case, the throughput of the server computer may be inadequate to meet the requested distribution rate. The solutions known in the art then result in slowing down the operation of the whole network.